I have been playing around with IBM’s Speech to Test Watson API
You need to sign up at Bluemix. Then log in.
Then start a new service: Watson, Speech to Text.
Manage your new Speech to Text service. Go to “Service Credentials.” Here get your username and password for the IBM Watson Speech to Text Service.
Use this command at your linux command line:
curl -u username:password -X POST –header “Content-Type: audio/wav” –header “Transfer-Encoding: chunked” –data-binary @myaudio.wav “https://stream.watsonplatform.net/speech-to-text/api/v1/recognize?continuous=true×tamps=true&max_alternatives=3” > out.txt